Extracting features with structural skeleton framework for semantic image classification by using supporting vector machine

Abstract

Searching images their semantic is an active problem in multimedia image retrieval. Many researchers have attempted to improve semantic models by using high-level concept based on keyword annotation. However, the annotation is tedious, in consistent, and erroneous. The retrieval process of such approaches is done by keyword searching. This model is rather rudimentary and it does not specific enough for representing the actual meaning. In this paper, we present a technique of the semantic image classification by using the human perception. The structural skeleton is used to extract the object components and image meaning. The feature selection methods are introduced to select the essential features from existing features. The experimental results indicate that our proposed approach offers significant performance improvements in the interpretation of semantic image classification, compare with other features, with the maximum of 93.80%